The measurement of water depth and sediment deposition is useful for understanding flow and siltation processes in pipes, channels, overland flows and tidal beaches. Sedimentation is of significant interest to geologists studying the evolution of the sea bed resulting from turbidity currents and underwater landslides. Conservationists monitoring the erosion and deposition of coastal beaches pay particular attention to the evolving behaviour of tidal levels and sediment transportation. Sediment build-up is also a major concern for water companies, particularly in sewer pipes where sediment build up can easily occur and can seriously reduce the capacity of the pipe while increasing the risk of surcharging during storm events. Therefore, it is desirable to measure both the flow depth and the amount of sediment present. Embodiments of the present invention enable the measurement of flow and sediment parameters in the context of a sewer pipe but it is to be understood that the invention is easily applicable to other areas of sediment transport and flow analysis.
Flow depth measurement can currently be accomplished by various means. Pressure sensors may be placed on the bed (for example in Bishop, Craig T. and Donelan, Mark A. 4, 1987, Coastal Engineering, Vol. 11, pp. 309-328), ultrasonic devices can be used above or within the flow to calculate the time for an acoustic signal to be reflected from the surface and hence its proximity (see, for example, U.S. Pat. No. 4,221,004) or even a simple visual scale protruding through the surface could be used. Perhaps the most all-round practical method is the use of conductance-based wave-probes. These devices consist of two parallel wires penetrating the water surface, and when a potential difference is applied across them an output voltage proportional to the conductance between the two wires and hence proportional to the submergence of the wires is measured. Similar to ultrasonic devices and pressure sensors, wave-probes may be operated remotely, and may be programmed to generate a warning if the water level surpasses a pre-determined safe range. Wave-probes offer a more economic alternative to ultrasonic devices and pressure sensors, with greater accuracy, and can also record a time series of surface level fluctuations. Similarly, capacitance-type wave probes, consisting of a single insulated wire, may be used to calculate the capacitance, which is a function of the depth and electrical properties of the water. However, wave-probes (conductance or capacitance type) must be calibrated to a specific location and configuration in order to output a meaningful depth measurement. Furthermore, any calibration may be affected by fluctuations in the local electrical conditions of the fluid such as the presence of impurities or pollutants in the fluid.
There are also a range of possible techniques for quantifying the depth of sediment at a particular point. Where possible, a simple visual scale can be used, but this is only usually an option for laboratory work where one can see the sediment level through the side of a tank or flume. Some electrical solutions have also been suggested as described below.
De Rooij et al. (de Rooij, F, Dalziel, S B and Linden, P F. 1999, Experiments in Fluids, Vol. 26, pp. 470-474) describe a system for electrically quantifying sediment layer thickness based on the measured resistance across the sediment layer. An array of electrodes is attached to the bed of a rectangular tank, and a reference electrode is positioned within the fluid. A change in sediment layer thickness produces a change in the resistance between the bed mounted electrodes and the reference electrode. Once calibrated, the measurement of the variation in the resistance enables accurate quantification of the sedimentation. There are however several limitations that present problems when attempting to implement the system in the field. The reference electrode is a cylindrical conductive rod which is required to be fully submerged. If the top of this electrode protrudes from the fluid surface then any surface fluctuations obscures the sedimentation measurement. Hence the technique is unsuitable for environments where the flow depth can vary significantly. Similarly, if the sediment layer forms to a depth in excess of the lower extremity of the reference electrode then the measurement is adversely affected. The most striking limitation of the system is that the measurement is only reliable while the system is still accurately calibrated, that is while the electrical properties of the fluid and the electrical and geometric properties of the sediment particles remain constant. As such this measurement technique is extremely effective for laboratory tests where the fluid properties are constant and sediment properties are more easily measureable, but not suitable for field measurements where the properties of the fluid and sediment are dynamic.
U.S. Pat. No. 5,032,794 (Ridd et al.) discloses a sediment measurement device consisting of a thin rod with ring electrodes positioned at intervals along its length. One electrode generates an electric field within the sediment layer and the other electrodes detect the voltage level at known positions in the electric field. Based on the voltage readings, the position of the sediment interface relative to the device may be calculated. The device is designed to be operated either in a horizontal orientation below the sediment layer, or in a vertical orientation penetrating the layer. The device is stated to be accurate to within 5% of the sediment depth, which is deemed acceptable for the context of coastal beach erosion, but is not suitable for detailed fluvial sediment analysis or pipe siltation processes, where the variation in sediment level can be very small but very significant. The device also has several other limitations. The fixed current source means that if measurements over a large range are required, the power supply must be suitably large. In the disclosure of Ridd et al. all embodiments of the device are mains powered. Although Ridd discloses that multiple boundaries may be detected by additional receiving electrodes, this is unfeasible if the physical separation of the receiving electrodes is too high (e.g. a high flow depth). The mathematical theory behind the device also assumes that there is an abrupt conductivity barrier to detect. The theory does not allow for (or quantify) gradual transitions in conductivity that are present in many sediment layers. Such a device may be practical for coastal environments where the sand sediment is reasonably uniform, but in most fluvial and sewer environments the sediment layer consists of gradual transitions in conductivity between coarse and fine sediment, and between closely packed and sparsely packed sediment. Transitions in sediment properties are of significant importance in understanding the composition and evolution of complex sediment layers and the device disclosed in Ridd et al. does not measure or identify such transitions. Similarly a more gradual sediment-flow interface caused by partially suspended sediment is difficult to detect. Furthermore, the theory is based on 3-dimensional electrical fields, and as such the device cannot function properly when other obstacles interfere with the expected field (for example the walls of a sewer pipe). Additionally, although Ridd's device gives a single sediment position reading at a point, it does not give an indication of the local gradient of the sediment layer, something very useful for understanding the erosion and deposition processes occurring. Perhaps most importantly the device is of relatively complex design and as such is financially and practically unsuitable for widespread deployment throughout a sewer system or even a coastal area. Finally, Ridd's device requires exposed conductors.
Jansen et al. (Jansen, Daniela, et al. 2005, Marine Geology, Vol. 216, pp. 17-26) disclose measuring the conductivity profile of the sea bed by lowering a conductivity sensor on a weight into the sediment. Pressure sensors calculate the depth of each conductivity reading and as such a conductivity profile plot is produced. This allows the detection of gradual changes in sediment layer composition but is unfeasible for small scale investigation such as in sewer pipes. In practice, this system is too complex to implement in small scale applications (for example in sewer systems, river reaches) and prohibitively costly for widespread use. Furthermore, the device only enables readings to be taken at discrete positions making the determination of sediment level accurate to the resolution of the measurement grid.
Similar to the depth measurement problem, quantification of sediment level is perhaps most easily achieved by conductance wave probes. Once calibrated, any change in sediment level may be indicated by the conductivity reading, since the wet sediment layer has different conductance characteristics to the water. Practically, however, this relies on the water surface remaining at a fixed level, something that is very unlikely to happen in most flow scenarios. Thus, the electrical and geometric properties of the sediment and the flow must remain constant for the calibration to remain true. This is not the case in a dynamic field environment.
Clearly there is a need for a device that can measure a conductivity profile (or profile of a different measurable electrical property), thereby locating the position of both the sediment level and the flow depth, and further measuring any changes in conductivity (or a different measurable electrical property) throughout the fluid and sediment layers. Wave probes currently provide a low cost, robust and simple measurement of water depth and surface fluctuation for a single position, but if the level of the bed changes due to sediment transport, this also affects the wave probe reading. Similarly, wave probes are also used to measure sediment position, but if the flow level changes then the calibration is invalidated. A device and method for measuring an accurate depth of fluid, depth of sediment and any changes to the position of the fluid and/or sediment would enable real time monitoring of sediment build up in sewer pipes while also providing a method and device to facilitate the monitoring of sediment erosion and deposition in a wide range of applications. Quantification of the conductivity (or a different measurable electrical property) throughout the sediment layer provides insight into the composition of the sediment, which is critical in predicting the mechanism of previous and future sediment deposition and erosion. Quantification of the fluid conductivity (or a different measurable electrical property) can give an indication of the presence of pollutants or of the degree of suspended sediment. Combining this with a measurement of local fluid and sediment surface gradients presents an unrivalled measurement device for use in many areas of science and ecology.
A single probe, comprised of two conductive wires of known dimensions may be used to measure the conductivity of a medium. Such a device, however, requires careful calibration with mediums of known conductivity. Single probe devices are also unable to ascertain the different contributions from several mediums of different conductivity and therefore are not useful for measuring fluctuations in sediment deposition and fluid level in mediums with varying conductivities. The same applies for other electrical properties such as capacitance and signal attenuation.
The conductivity of a solution is simply a measurement of the quantity of ions (charged atoms or molecules) in the fluid. The more ions present in the fluid, the more conductive the fluid becomes, since there are more charge carriers which enable a greater flow of electric charge. Electric current in an electrolyte is the flow of ions between the two electrodes.
Conductivity is the ability of a material to conduct electric current. The method used to measure conductivity is simple: two wires are placed in the sample, a potential is applied across the wires (normally a sine wave voltage), and the current is measured. Conductivity (G), the inverse of resistivity (R), is determined from the voltage and current values according to Ohm's law (and is also a function of the probe wire geometries, which are designed to remain constant).